US 11,514,363 B2 · Using a recursive reinforcement model to determine an agent action

Abstract

According to examples, an apparatus may include a processor and a memory on which is stored machine readable instructions that may cause the processor to access data about an environment of an agent, identify an actor in the environment, and access candidate models, in which each of the candidate models may predict a certain action of the identified actor. The instructions may also cause the processor to apply a selected candidate model of the accessed candidate models on the accessed data to determine a predicted action of the identified actor and may implement a recursive reinforcement learning model using the predicted action of the identified actor to determine an action that the agent is to perform. The instructions may further cause the processor to cause the agent to perform the determined action.

Claims20

Dependent claims

An apparatus comprising:

a processor;
and
a memory on which is stored machine readable instructions that cause the processor to:
access data about an environment of an agent;
identify an actor in the environment;
access candidate models, wherein each of the candidate models predicts a certain action of the identified actor;
apply a selected candidate model of the accessed candidate models on the accessed data to determine a predicted action of the identified actor;
implement a recursive reinforcement learning model using the predicted action of the identified actor to determine an action that the agent is to perform;
and
cause the agent to perform the determined action.

7 more elements

A method comprising:

accessing, by a processor, data about an environment of an agent;
identifying, by the processor, a first actor and a second actor in the environment;
applying, by the processor, a first order reinforcement learning model on the second actor to determine a first order predicted action of the second actor;
applying, by the processor, a second order reinforcement learning model on the first actor to determine a second order predicted action of the first actor, the second order reinforcement learning model using a predicted reward of the first order predicted action to determine the second order predicted action of the first actor;
applying, by the processor, a third order reinforcement learning model on the agent to determine an action that the agent is to perform, the third order reinforcement learning model using a predicted reward of the first predicted action to determine the action that the agent is to perform;
and
causing, by the processor, the agent to perform the determined action.

4 more elements

Dependencies are inferred from the claim text.

Using a recursive reinforcement model to determine an agent action

Life of this patent

Using a recursive reinforcement model to determine an agent action

Life of this patent

Abstract

Claims20